STAT545 hw05

Setting ups

suppressPackageStartupMessages(library(tidyverse))
suppressPackageStartupMessages(library(gapminder))
suppressPackageStartupMessages(library(forcats))
suppressPackageStartupMessages(library(scales))
suppressPackageStartupMessages(library(plotly))

Part 1: Factor management

First we need to ensure which of the columns are factors. From the output of str function we can see here country and continent are factors. The number of levels in country is 142, and 5 for the continent.

gapminder %>% 
  str()

## Classes 'tbl_df', 'tbl' and 'data.frame':    1704 obs. of  6 variables:
##  $ country  : Factor w/ 142 levels "Afghanistan",..: 1 1 1 1 1 1 1 1 1 1 ...
##  $ continent: Factor w/ 5 levels "Africa","Americas",..: 3 3 3 3 3 3 3 3 3 3 ...
##  $ year     : int  1952 1957 1962 1967 1972 1977 1982 1987 1992 1997 ...
##  $ lifeExp  : num  28.8 30.3 32 34 36.1 ...
##  $ pop      : int  8425333 9240934 10267083 11537966 13079460 14880372 12881816 13867957 16317921 22227415 ...
##  $ gdpPercap: num  779 821 853 836 740 ...

gapminder$continent %>%
  levels()

## [1] "Africa"   "Americas" "Asia"     "Europe"   "Oceania"

gap_wo_oc <- gapminder %>%
  filter(continent != "Oceania") %>%
  droplevels()

gap_wo_oc %>%
  str()

## Classes 'tbl_df', 'tbl' and 'data.frame':    1680 obs. of  6 variables:
##  $ country  : Factor w/ 140 levels "Afghanistan",..: 1 1 1 1 1 1 1 1 1 1 ...
##  $ continent: Factor w/ 4 levels "Africa","Americas",..: 3 3 3 3 3 3 3 3 3 3 ...
##  $ year     : int  1952 1957 1962 1967 1972 1977 1982 1987 1992 1997 ...
##  $ lifeExp  : num  28.8 30.3 32 34 36.1 ...
##  $ pop      : int  8425333 9240934 10267083 11537966 13079460 14880372 12881816 13867957 16317921 22227415 ...
##  $ gdpPercap: num  779 821 853 836 740 ...

gap_wo_oc$continent %>%
  levels()

## [1] "Africa"   "Americas" "Asia"     "Europe"

big_pop <- gapminder %>% 
  filter(continent == "Americas") %>%
  group_by(country) %>%
  summarise(pop = pop[year == 2007]) %>%
  filter(pop > 1e7)

big_pop %>%
  mutate(country = fct_reorder(country, pop)) %>%
  ggplot(aes(pop, country)) +
  geom_point()

big_pop_ordered <- big_pop %>%
  arrange(pop)

knitr::kable(big_pop_ordered)

big_pop_ordered %>%
  ggplot(aes(pop, country)) +
  geom_point()

big_pop_reordered <- big_pop %>%
  mutate(country = fct_reorder(country, pop)) %>%
  arrange(pop)

knitr::kable(big_pop_reordered)

big_pop_reordered %>%
  ggplot(aes(pop, country)) +
  geom_point()

Part 2: File I/O

In this part we can use the previous big_pop data frame and do some file IO experiments. First we save the data frame using write_csv, write_tsv, saveRDS and dput.

big_pop_saved <- tail(big_pop_reordered, 6) %>%
  droplevels()

knitr::kable(big_pop_saved)

write_csv(big_pop_saved, "big_pop.csv")
write_tsv(big_pop_saved, "big_pop.tsv")
saveRDS(big_pop_saved, "big_pop.rds")
dput(big_pop_saved, "big_pop.txt")

Then we read them back and see what happens.

big_pop_csv <- read_csv("big_pop.csv")

## Parsed with column specification:
## cols(
##   country = col_character(),
##   pop = col_integer()
## )

knitr::kable(big_pop_csv)

big_pop_tsv <- read_tsv("big_pop.tsv")

## Parsed with column specification:
## cols(
##   country = col_character(),
##   pop = col_integer()
## )

knitr::kable(big_pop_tsv)

big_pop_rds <- readRDS("big_pop.rds")
knitr::kable(big_pop_rds)

big_pop_txt <- dget("big_pop.txt")
knitr::kable(big_pop_txt)

Part 3: Visualization design

The graph below shows where we start. It is not too bad, but with a little more working we can make it different.

gapminder %>% 
  ggplot(aes(gdpPercap, 
             lifeExp, 
             color = year)) + 
  geom_point() + 
  scale_x_log10()

Now here is what it looks like after some hard working. It is more readable, with more infos, and has a title now so people know what it is for.

(p <- 
  gapminder %>% 
  filter(continent != "Oceania") %>%
  ggplot(aes(gdpPercap, lifeExp)) + 
  geom_point(aes(color = year), alpha = 0.4) + 
  scale_x_log10(labels = dollar_format()) +
  scale_colour_distiller(palette = "Greens") +
  facet_wrap(~continent) + 
  scale_y_continuous(breaks = 10*1:10) +
  theme_bw() +
  labs(x = "Gdp per capita",
       y = "Life expectency",
       title = "Life expectency against gdp per capita of four continents"
       ) +
  theme(axis.text = element_text(size = 14),
        strip.background = element_rect(fill = "orange"),
        panel.background = element_rect(fill = "gray")
        )
)

Then we can trun this into a plotly graph. It allows more interactions with the graph compared with ggplot graph, including zoom in, select regions you want to see, or check the value of every point in the plot. It is cool and more suitable for presentations!

ggplotly(p)

## We recommend that you use the dev version of ggplot2 with `ggplotly()`
## Install it with: `devtools::install_github('hadley/ggplot2')`

Part 4: Writing figures to file

By using ggsave we can save the plot to a local file.

ggsave("plot.png", p)

## Saving 7 x 5 in image

Then we can load it back from files in a html way.

We can actually save them in different format and different quality.

ggsave("plot_l.bmp", p, device = "bmp", height = 5, width = 5, dpi = 50)
ggsave("plot_h.bmp", p, device = "bmp", height = 5, width = 5, dpi = 200)

Then we can load them back and see the difference.